Angioplasty and Stenting for Restenosis After Carotid Endarterectomy
Background and Purpose Endarterectomy for recurrent carotid stenosis after endarterectomy has a significantly higher complication rate than the original operation. Angioplasty and stenting may offer a useful alternative treatment for these patients.
Methods Between September 1994 and April 1996, 22 patients had 25 carotid arteries treated with angioplasty and stenting for postendarterectomy restenosis. All patients had an independent neurological examination and National Institutes of Health Stroke Scale evaluation before and after the procedure. Patients were treated with aspirin and ticlopidine. All patients were requested to return at 6 months for follow-up angiography. The mean patient age was 69±7 years, and the mean elapsed time from endarterectomy was 73±69 months. Seventy-seven percent of the patients were symptomatic.
Results Mean stenosis was reduced from 79±13% before the procedure to 1.8±3.6% after stenting. One patient had a minor stroke, for a complication rate of 4% per treated artery. In the eight patients who returned for 6-month angiography, mean stenosis was 19.4±4.4% and restenosis (≥50% stenosis) did not occur.
Conclusions In a small series, angioplasty and stenting appear to be safe and well tolerated for the treatment of postendarterectomy restenosis.
Recent randomized trials have shown CEA to be an effective treatment that reduces the risk of stroke in symptomatic and asymptomatic patients with severe extracranial carotid stenosis.1 2 Although complete angiographic follow-up studies have not been done after CEA, restenosis appears to occur in 8% to 19% of patients.3 4 5 6 7 8 The scarring from the initial operation makes repeated CEA difficult, and some experienced surgeons have reported higher complication rates for repeated operations than for primary CEA.9 10 On theoretical grounds, angioplasty has been forwarded as an effective treatment option for these patients, but a well-documented series has not been reported. We report a consecutive, prospectively evaluated series of patients with restenosis after CEA who were treated with carotid angioplasty and stenting.
Subjects and Methods
Between September 1994 and April 1996, 151 patients had angioplasty and stenting of 181 carotid arteries at the University of Alabama Hospital. Of these patients, 22 had angioplasty and stenting of 25 carotid arteries for restenosis after CEA and form the basis for the study population. Inclusion criteria were based on those of the NASCET and Asymptomatic Carotid Atherosclerosis Study, which specified that patients with symptomatic or asymptomatic carotid stenosis of at least 60% were eligible. Stenoses were measured according to NASCET methodology with the distal, normal vessel serving as the reference diameter.1 Exclusion criteria were similar to those of NASCET with several important differences.11 Patients were not excluded for the following: (1) severe distal internal carotid stenosis that is surgically inaccessible; (2) age older than 79 years; (3) heart, kidney, liver, or lung failure or cancer likely to cause death within 5 years; (4) nonatherosclerotic stenosis; (5) cardiac valvular lesion or rhythm disorder likely to be associated with cardioembolic stroke; (6) previous ipsilateral CEA; (7) unstable angina or a myocardial infarction within the previous 6 months; (8) progressing neurological signs; (9) history of contralateral CEA within 4 months; or (10) history of a major surgical procedure within the last 30 days.
Patients were excluded only for the following reasons: presence of an intracranial tumor or arteriovenous malformation; severe disability from stroke or dementia; intracranial stenosis that exceeded the severity of the extracranial stenosis; severe aortic or peripheral vascular disease that precluded vascular access; severe renal insufficiency but not yet on dialysis; and inability to give informed consent. Of the patients referred to us with post-CEA restenosis, none were excluded. The study protocol was approved by the institutional review board of the University of Alabama at Birmingham Hospital.
Clinical and Imaging Protocol
A complete neurological history and examination were performed on all patients by an experienced neurologist. An independent neurologist not involved in the interventional procedure evaluated patients using the NIH Stroke Scale before the procedure, at 24 hours after the procedure, at 6 weeks after the procedure, and at 6 months.12
All patients were categorized according to the Mayo Clinic CEA risk classification scheme: grade 1, no significant risk factors; grade 2, angiographic risk factors; grade 3, medical risk factors such as severe coronary artery disease; grade 4, unstable neurological syndromes; grade 5, acute carotid occlusion; and grade 6, previous CEA.3
MRI or CT of the head and complete diagnostic cerebral angiography were performed on all patients. The MRI and angiograms were reviewed by an experienced neuroradiologist. If a patient had neurological deterioration after angioplasty and stenting, MRI or CT of the head was repeated. All patients were requested to have follow-up angiography at 6 months.
Carotid Angioplasty and Stenting Protocol
Aspirin (1 tablet QD) and ticlopidine (250 mg BID) were given for at least 2 days before the procedure as well as on the morning of the procedure. Heparin (5000 U) was administered intravenously; further boluses were given as needed to maintain the activated clotting time during the procedure at 200 to 250 seconds. Frequent neurological checks were performed by a neurologist during the procedure. Femoral venous access was gained in all patients, and a pacemaker was either placed in the right ventricle or made immediately available. Atropine (1 mg) was given for bradycardia. Blood pressure was carefully monitored and modulated with boluses of metaraminol or intravenous nitroglycerin as required.
A 9F sheath was placed in the femoral artery and the common carotid artery catheterized with a 5F Newton catheter (USCI) over a 0.038-inch hydrophilic coated wire (Glidewire, Medi-tech; Roadrunner, Cook Inc). The diagnostic catheter was exchanged over an 0.038-inch extra-stiff Amplatz wire (Cook Inc) for a 9F multipurpose large-lumen guiding catheter (Cordis Inc). On-line quantitative angiography was performed, with the contrast-filled guiding catheter for calibration, to precisely measure the stenosis and the diameter of the distal internal carotid artery and accurately size the balloons and stents required. With the use of a standard coaxial system, the lesion was crossed with flexible coronary guidewires. The size of the initial angioplasty balloon was dictated by the severity of the stenosis. Very severe lesions were predilated with low-profile coronary balloons; in the case of less severe lesions, the initial dilatation might be performed with a definitive balloon sized to the distal normal artery. A stent was then deployed across the lesion. The stent was further dilated at high pressure (14 to 16 atm) to firmly embed it into the vessel wall. Completion angiography was performed on the ipsilateral intracranial vessels.
The sheaths were removed later that day; poststent anticoagulation was not used. Patients were discharged on either the first or second day after the procedure. Ticlopidine was continued for 3 weeks, and aspirin was continued permanently. Ticlopidine has been found to decrease stent thrombosis.13
Data Collection and End Points
All clinical, angiographic, and stenting data were prospectively recorded on standard forms by a physician and research nurse coordinators. Clinical and laboratory details were recorded continuously during the hospitalization. The primary clinical end points were as follows: (1) any minor or major stroke, myocardial infarction, or death; (2) ipsilateral minor or major stroke; and (3) target lesion revascularization. Angiographic end points were as follows: (1) minimal luminal diameter and percent stenosis after stenting; (2) angiographic success rate, defined as achieving a less than 30% residual stenosis; and (3) minimal luminal diameter and percent stenosis on the follow-up angiogram at 6 months.
A TIA was defined as a new neurological deficit that resolved completely within 24 hours. A minor stroke was defined as a new neurological deficit that either resolved completely within 7 days or increased the NIH Stroke Scale score by 3 or less. A major stroke was defined as a new neurological deficit that persisted after 7 days and increased the NIH Stroke Scale score by 4 or more. Myocardial infarction was defined as the development of new pathological Q waves or elevation of creatine kinase to more than twice the normal range with an elevated MB fraction. Target lesion revascularization was defined as all surgical or percutaneous revascularization involving the target lesion, including repeated angioplasty or stenting, coronary artery bypass, CEA, or EC-IC arterial bypass.
All values are expressed as mean±1 SD.
Patient characteristics are summarized in Table 1⇓, and case information is provided in Table 2⇓. The mean elapsed time from CEA to presentation with restenosis was 73±69 months (range, 1 to 223 months). Seventy-seven percent of the patients were symptomatic, and 80% of the treated arteries had caused symptoms.
All procedures were angiographically successful. Stenosis before the procedure was 79±13%, and the minimal luminal diameter was 1.6±1.1 mm. After stenting, residual stenosis was 1.8±3.6%, and the minimal luminal diameter was 4.6±1.1 mm.⇓ Thirty-six stents were deployed in 25 arteries. Twenty of the stents were Johnson & Johnson biliary stents (Johnson & Johnson Interventional Systems; Fig 1), 14 were Schneider Wallstents (Schneider Inc; Fig 2), and 2 were Cook flexible stents (Cook Inc). Three patients had bilateral carotid stenting during a single procedure. Two patients had stenting of a vertebral artery during the carotid artery stenting procedure, 1 patient had stenting of both iliacs and the aorta during the carotid stenting procedure, and 1 patient was sent to EC-IC bypass for a contralateral carotid occlusion.⇓
One periprocedural stroke occurred. Patient 7 (Table 2⇑) had a minor stroke 48 hours after angioplasty and stenting. Review of his angiographic films revealed a small distal dissection that was left uncovered by the stent. He was brought back to the angiography suite, and another stent was placed distally to cover the dissection. One patient had a groin hematoma that required repeated hospitalization for observation but did not require intervention or blood transfusion. The mean length of stay was 1.7±1.9 days. The mean NIH Stroke Scale score was 2.2±4.5 before the procedure and 1.9±4.5 24 hours after the procedure. One patient had a significant improvement in his baseline hemiparesis within 24 hours of the angioplasty and stenting.
The mean length of follow-up was 8 months. There were no clinical events during the follow-up period. Eight patients have reached the 6-month angiographic end point. The mean follow-up angiographic stenosis was 19.4±4.4% (range, 13% to 26%), and the minimal luminal diameter was 4.3±0.5 mm. There have been no repeat angioplasty or CEA procedures.
The primary cause of death in patients with peripheral vascular disease is coronary artery disease. As the detection and treatment of coronary artery disease have improved, patients with peripheral vascular disease survive longer and recurrent carotid stenosis after CEA is encountered more frequently. This carefully conducted prospective observational study suggests that percutaneous angioplasty and stenting after CEA is a safe and effective therapeutic option.
There appear to be several different causes of recurrent carotid stenosis after surgery.10 The two most frequent causes are myointimal hyperplasia and progressive atherosclerosis. Myointimal hyperplasia is a normal response after CEA, and its exuberance varies from patient to patient. Whether it is related to surgical technique, such as the depth to which the media is removed during the CEA, is unclear. Myointimal hyperplasia appears to be the primary cause of restenosis within the first 24 months after CEA. Progressive atherosclerosis plays a greater role in more delayed restenosis.
Other significant causes of restenosis include organized thrombus and scarring at the proximal and distal ends of the CEA. Scarring appears to be more frequent in the common carotid than in the distal internal carotid artery. Controversy remains as to whether vein patch closure results in a lower incidence of restenosis than primary closure.3 9 Risk factors for restenosis may include age less than 60 years, female sex, primary closure, continued smoking, diabetes, and hypercholesterolemia.6 9 10 14 16 We cannot be certain of the cause of restenosis in our patients on the basis of the angiographic appearance alone. Five of our patients presented with restenosis within 12 months of CEA, and the presumptive cause of restenosis in these patients would have to be myointimal hyperplasia. One of these four patients had CEA only 1 month previously and had a very tight bandlike stenosis at the distal CEA site. This could represent exuberant myointimal hyperplasia or a residual lesion. Seventeen of our patients had undergone the procedure more than 12 months previously and probably had recurrent atherosclerosis. It has been hypothesized that myointimal hyperplasia has a lower risk of stroke than atherosclerotic disease, but all of our patients who presented within 12 months of ipsilateral CEA had a TIA or stroke.
It is difficult to accurately estimate the incidence of restenosis after CEA. Several longitudinal studies of CEA have been published but are limited by either small numbers of patients, retrospective study designs, or limited ultrasound follow-up: 41% to 76% of patients in several large series.7 8 14 In a careful prospective study from the Mayo Clinic with serial intravenous digital subtraction angiography, clinical follow-up was 97% complete but digital subtraction angiography follow-up was accomplished in only 66% of patients. Stenosis was mild to moderate in 10% of the these follow-ups and severe (>70%) in 3%.3
Similarly, it is difficult to reach a conclusion regarding the prognosis of recurrent carotid stenosis. In the Cleveland Clinic series of carotid reoperations, 5.1% of the patients had neurological symptoms leading to the repeated operations.9 In the Mayo Clinic series of reoperations, 90% of the patients were symptomatic.10 Since these studies represent repeated interventions, there is a case-selection bias toward symptomatic patients. In a retrospective study with limited follow-up, 4 (22%) of 18 patients with restenosis developed neurological symptoms.6 In another retrospective study in which 224 of 544 endarterectomies (41%) underwent serial ultrasound examinations, the incidence of restenosis was 12.7%, with 23% being symptomatic.8 In a retrospective study in which 78 of 143 endarterectomies underwent serial ultrasound follow-up, 14 (18%) had restenosis and only 2 (14%) of these lesions were symptomatic.7
Repeated operation is significantly more difficult than the initial operation,10 largely because of the dense scarring both outside and inside the carotid artery and difficulty in obtaining tissue cleavage plains. The scarring external to the artery lengthens the duration of dissection and generally requires more manipulation of the carotid bifurcation, with an attendant increase in the risk of embolization.10 The myointimal hyperplasia inside the artery blends with the normal intima, making it difficult to obtain a cleavage plane. In some cases CEA cannot be performed, and the involved segment must be excised and an interposition graft placed. Given the technical challenges of a repeated operation, it is not surprising that the complication rate is higher. The major complication rate for repeated carotid operation for the Mayo Clinic and Cleveland Clinic series ranges from 4.6% to 10.9%.9 10 Some retrospective series have, however, reported complication rates that are no higher than for primary endarterectomy. Coyle and coworkers17 reported a 1.4% stroke rate in 69 repeated operations, and Gagne et al18 reported no strokes in 47 repeated operations.
In the present series, major complications were avoided and there was one minor stroke, for a complication rate of 4% per treated artery. This minor stroke occurred early in our experience and was due to a technical error. Since an endovascular approach was used, the internal and external scarring at the CEA site does not pose a problem for angioplasty and stenting. In addition, many of these lesions are quite distal (at the distal end of the endarterectomy) or quite proximal (in the common carotid) and therefore are difficult to reach surgically. In the follow-up period, neither restenosis nor neurological events occurred.
It is important to note that this series of patients represents the developmental phase of a new technique. These procedures were performed with equipment not specifically designed for the carotid arteries. As more experience is gained with carotid stenting and equipment is specifically designed for this application, the complication rate should decrease further. This study is limited by the small number of patients and limited duration of follow-up. Work is ongoing to expand the patient population and determine whether these results can be replicated in a multicenter experience. In this challenging group of patients, the initial results are encouraging and warrant further investigation of endovascular intervention for recurrent carotid stenosis after CEA.
Selected Abbreviations and Acronyms
|NASCET||=||North American Symptomatic Carotid Endarterectomy Trial|
|NIH||=||National Institutes of Health|
|TIA||=||transient ischemic attack|
The authors wish to acknowledge Ronald Levine, MD, Christopher Goods, MD, Suresh P. Jain, MD, Khaled Al-Shaibi, MD, and Atul Mathur, MD, for their assistance with procedures and data collection.
- Received May 10, 1996.
- Revision received July 22, 1996.
- Accepted July 22, 1996.
- Copyright © 1996 by American Heart Association
Meyer FB, Piepgras DG, Sundt TM Jr, Fode NC, Marsh WR. Recurrent carotid stenosis. In: Meyer FB, ed. Sundt's Occlusive Cerebrovascular Disease. Philadelphia, Pa: WB Saunders Co; 1994:310-321.
North American Symptomatic Carotid Endarterectomy Trial (NASCET) Steering Committee. North American Symptomatic Carotid Endarterectomy Trial: methods, patient characteristics, and progress. Stroke. 1991;22:711-720.
Brott T, Adams HP Jr, Olinger CP, Marler JR, Barsan WG, Biller J, Spilker J, Horleran R, Eberle R, Hertzberg V. Measurements of acute cerebral infarctions: a clinical examination scale. Stroke. 1989;20:864-870.
Schomig A, Newmann FJ, Kastrati A, Schuhlen H, Blasini R, Hadamitzky M, Walter H, Zitmann-Roth EM, Richardt G, Alt E, Schmitt C, Ulm K. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary-artery stents. N Engl J Med.. 1996;334:1084-1089.
Coyle KA, Smith RB, Gray BC, Salam AA, Dodson TF, Chaikof EL, Lumsden AB. Treatment of recurrent cerebrovascular disease: review of a 10-year experience. Ann Surg.. 1995;5:517-521.